Professor Andrew Biankin from the University of Glasgow explains how Glasgow has become a centre of excellence for precision oncology research in the UK.
Half of the UK’s top dozen pancreatic cancer researchers are found at the University of Glasgow. They work at the Cancer Research UK (CRUK) Glasgow Centre, a partnership between the Beatson West of Scotland Cancer Centre, the CRUK Beatson Institute, NHS (National Health Service) Greater Glasgow & Clyde, and the Universities of Glasgow and Strathclyde, that brings together scientists and clinicians to develop the very best in cancer research, drug development and patient care. While pancreatic cancer is a major strength, research is also conducted in colorectal, ovarian, prostate and brain cancer, chronic myeloid leukaemia and other cancers.
In March 2017, CRUK invested £10M in Precision-Panc, a purpose-built platform to develop new treatments for pancreatic cancer led by researchers at the University of Glasgow and the CRUK Beatson Institute, in collaboration with scientists and clinicians across the UK including Cambridge, Manchester, Oxford and the Institute of Cancer Research in London. This investment in Precision-Panc supports clinical trials delivered through the nation-wide National Institute for Health Research-NHS network, preclinical work, assay development, biomarker work and a large amount of molecular sequencing and is a paradigm for the development of precision oncology towards a self-learning healthcare system that can be used as an exemplar for other cancer types.
For precision oncology to achieve its potential, there is a need to amass large datasets of deep molecular and detailed phenotypic data through healthcare systems as part of routine care. A learning healthcare system is characterised by the continuous generation and implementation of knowledge from clinical trials ‘embedded’ within the real world of healthcare delivery.
A precision oncology learning healthcare system aligns molecular science, informatics, incentives and culture, and utilises existing clinical staff and systems to embed clinical testing with novel therapies within the healthcare delivery process. In this way, new knowledge is captured as an integral component of healthcare delivery, supplying researchers with a constant accumulation of high quality molecular and clinical data.
As a demonstration of this concept, Precision-Panc clinical trials for patients with pancreatic cancer are currently embedded in the NHS at more than 20 sites nationally, with the number of centres running the trials constantly increasing. The aim is to offer the option of a clinical trial to as many pancreatic cancer patients as possible and to demonstrate the benefits and cost effectiveness of the model for this, and potentially for other cancers.
A high mortality cancer with no true ‘standard-of-care’
Pancreatic cancer has the worst outcome of any cancer, with a five-year survival of less than 5 percent – a mortality rate that has not improved for 50 years. Despite it being relatively uncommon – it is only the 11th most common cancer – it will soon become the second leading cause of cancer death in western society after lung cancer.
Most patients have widespread disease at the time of diagnosis and only around 15 percent of patients can have the cancer surgically removed. Even in those lucky enough to have surgery, some 80 percent will still die within five years.
Treatment with Gemcitabine, the standard of care for many years, either alone or more recently in combination with Erlotinib or Capecitabine, has had minimal impact on the disease, with the overall median survival still only around six months from diagnosis.
More recently, the Gemcitabine + nab-Paclitaxel and FOLFIRINOX combinations, which have significant toxicity, have become the standard of care for patients with good performance status. Yet, these only extend survival to a meagre 8.5 and 11 months respectively. Regrettably, the two-year survival for advanced pancreatic cancer remains less than 10 percent, presenting the argument that an adequate “standard of care” does not truly exist.
Importantly, small subgroups of patients who cannot yet be identified before treatment, derive significant benefit from some of these therapies. Identifying patients who will benefit before commencing treatment, not giving treatments to patients when they will not work and offering them instead innovative treatments in clinical trials may seem obvious but this is the central challenge facied by clinicians involved in cancer care.
Tackling the diversity of pancreatic cancer
Recent genomic studies have revealed pancreatic cancer to be composed of numerous molecular subtypes. These subtypes are often vastly different, with different clinical patterns of behaviour and therapeutic response. These differences cannot be identified under the microscope.
Now that scientific advances have enabled us to measure the molecular changes that underlie differences in disease patterns and drug response, and we can measure the differences with extreme accuracy, it is not surprising that we see a great degree of diversity. But while we know more about pancreatic cancer, we have not yet defined how to respond to these differences with better treatment approaches.
Another feature that makes pancreatic cancer particularly challenging to study and treat is that it is a less common cancer. Research and therapeutic development for the most common cancers with the highest numbers of patients can be studied effectively in single research institutions. Pancreatic cancer, however, is not common enough for all institutions to develop significant expertise, and it is vital that research is conducted collaboratively between networked institutions.
Our newfound understanding of the molecular diversity of disease including cancer presents a further challenge to the current system. Most therapies tested in pancreatic cancer have been developed for other cancer types. While many targets for these drugs exist in pancreatic cancer, they can occur at such low prevalence that testing in pancreatic cancer leads to ‘negative trials’, where the drug seems to have no benefit in unselected patient groups. Yet, for the right patient, maybe as few as 1 percent, the treatment can have an exceptional effect on their cancer.
In order to test targeted agents in groups of patients that are individually small, but collectively large, we needed to develop platforms to molecular test the patients so they can be allocated to appropriate trials where we think the drugs will work. The need to deliver this kind of precision oncology approach has seen the creation of programmes for many cancer types globally – US National Cancer Institute’s MATCH and Cancer Research UK’s Stratified Medicine Programme, Lung-Matrix. Precision-Panc is a networked, dynamic, therapeutic development platform that has been developed to address this need for pancreatic cancer.
A purpose-built platform
We have developed Precision-Panc to accelerate the evolution of pancreatic cancer treatment into the molecular age of oncology, so that biopsy, molecular directed care and therapeutic development become part of routine clinical practice in healthcare systems.
Precision-Panc is a synergistic and dynamic platform, aligning and coordinating preclinical discovery and clinical development, that delivers clinical trials through the NHS. Ultimately, we aim to offer molecular profiling to all patients with pancreatic cancer, along with viable clinical trial options, to find ‘the trial for the patient’, rather than ‘the patient for the trial’. Working within the world’s largest united healthcare system, the NHS, Precision-Panc creates opportunities for scientific research that were not previously possible and enables forward and backward translation between the laboratory and the clinic to advance scientific discovery and therapeutic development for precision oncology.
The principle is that integrative analysis of large multidimensional datasets identifies novel pathophysiology and potential improved treatment approaches for individually small, but cumulatively large, proportions of patients. Next-generation model systems that are extensively molecularly characterised and known to reflect important pancreatic cancer subtypes provide unique and powerful models to understand function and mechanism, to define and refine therapeutic strategies and evaluate biomarkers prior to assessment in human clinical trials.
Precision-Panc‘s PRIMUS (Pancreatic Cancer Individualised Multi-arm Umbrella Study) family of clinical trials, embedded in the NHS, drives the overall process, offering patients viable trial options. Also, by partnering with existing clinical trials and attracting new externally developed studies, the Precision-Panc framework can supply molecularly profiled participants for any clinical trial.
The framework enables scientific opportunities that were not previously possible by using serial, pre-and post-treatment examination of tissue and blood samples of pancreatic cancer patients on trial. Clinical trial and biospecimen data, in turn, provide additional information to continue to build an integrated pancreatic cancer databank to advance the discovery of novel vulnerabilities and treatment strategies.
The specific challenge in pancreatic cancer is that the available therapies only provide incremental overall benefit at best, leading to difficult decisions in selecting a treatment from an increasing number of available therapies that are not very effective overall, but may elicit an exceptional response in a specific individual. Unfortunately, there are no means to predict which individuals will benefit from which treatment options. With current treatments, the average benefit for all patients is measured in weeks to months, barely shifting overall survival rates. Nonetheless, when through happenstance a patient receives the most appropriate therapy, the results can be staggering.
With incremental overall improvements, and dramatic responses in yet undefined subgroups, we are approaching a problem – we need to give current standard therapy even though there is little benefit to most, because a few patients may respond dramatically. Whilst this may be possible in chronic disease or other cancers where the mortality is not as great and not as rapid, in pancreatic cancer this is challenging. This causes the added problem of a cul-de-sac of therapeutic development in that, by the time the patients have gone through an accumulating potential list of therapies with incremental benefit, they no longer have the opportunity of clinical trials.
There is excellent evidence that participation in clinical trials is associated with better outcomes for patients and for pancreatic cancer patients’ clinical trials with novel therapeutics can be their best treatment option, especially where current therapies are mostly ineffective. Precision-Panc’s PRIMUS trials approach pancreatic cancer patients early in the patient journey with the offer of a viable clinical trial as one of their treatment options. The single, serial approach to clinical trials where a trial is considered after all standard-of-care is exhausted, is usually too late for pancreatic cancer patients. Performing biopsies in series takes more time than most patients have, so it is much better to test for all potential options in parallel and have viable choices in a timely manner.
Glasgow Precision Oncology Laboratory
The University of Glasgow’s Glasgow Precision Oncology Laboratory (GPOL) supports the Precision-Panc platform with molecular testing of patient samples and performing laboratory experiments to develop better treatments. GPOL was established in 2016 to advance novel therapeutic strategies for cancer for research and industry. GPOL’s purpose-built capacity, expertise, infrastructure and resources are aligned to accelerate therapeutic development from discovery through to implementation in the clinic.
While GPOL is a world leader in pancreatic cancer, it also has expertise with other cancer types including colorectal, ovarian, brain, gastric and liver cancers, with an affordable, bespoke R&D service that is structured to combine research expertise in rapid whole-genome sequencing, bioinformatics, and patient pathways to precision oncology in real-world healthcare. GPOL has extensive experience of partnering with researchers, industry and the NHS to develop and implement national and international clinical trials.
GPOL’s capabilities include tests for discovery work on patient tumours, such as whole-genome sequencing and RNA-seq, and assessing epigenetic alterations, such as methylation. GPOL also conducts experiments to understand models of disease, particularly in genetically engineered mouse models, through defining their genomic mutational landscapes and transcriptome and then providing molecular phenotypes for clinical trials development.
GPOL’s Clinical Cancer Genome assay is designed to be ‘real-world compatible’, recognising that most clinical cancer samples are collected as formalin-fixed paraffin-embedded (FFPE) blocks. By accommodating this, GPOL’s assays maximise the number of patients who will have accessible samples for a trial and lower the barrier to subsequent clinical adoption.
GPOL’s sequencing workflow includes an Agilent Bravo Automated Liquid Handling Platform for library preparation automation, and Illumina’s Novaseq, HiSeq4000, and NextSeq instruments for next generation sequencing. This combination gives the capacity to turn around samples quickly, as some patients, especially those with aggressive high mortality cancers, have a limited amount of time for participating in clinical trials.
Through the University of Glasgow, GPOL also offers clinician training, PhD and Masters’ programmes.
Once the patient enters Precision-Panc, they are tested using the GPOL Clinical Cancer Genome assay and their multidisciplinary team receives a report showing the underlying molecular profile of their cancer, with relevant mutations and genomic events. The patients are offered choices of clinical trials that are relevant to their molecular profile, location and preferences. The patient and their treating clinician can then decide to participate in a trial or elect for standard care.
Longitudinal data is acquired not just for clinical trials, but also for standard treatment, and these are collated into the knowledge bank that continues to inform discovery.
Patient input for Precision-Panc
The involvement of people affected by pancreatic cancer is crucial to every stage of Precision-Panc. The Precision-Panc Patient-Public Involvement working group, comprising Precision-Panc staff and research nurses, and colleagues from Cancer Research UK, and patient organisations Pancreatic Cancer UK and Pancreatic Cancer Scotland, routinely consults and collaborates with patients and families/advocates to add value at all stages of the research cycle.
This group works to increase the quality of the research programme and to ensure that the aims and methods of the research are relevant, acceptable and feasible to people with pancreatic cancer; that the information around providing tissue samples and taking part in any eventual precision medicine clinical trial is clear and accessible; and that the outcomes of the study will have an impact on the way pancreatic cancer is understood amongst patients.
The seven-member Precision-Panc Patient Advisory Group is recruited from Pancreatic Cancer UK’s Research Involvement Network of more than 120 patients, survivors, carers and family members and Cancer Research UK’s Your Involvement Network of more than 900 people affected by different types of cancer, including those with personal cancer experience, carers or family members.
The Precision-Panc Patient Advisory Group have representatives on the Precision-Panc Steering Committee, Scientific Advisory Board and Umbrella Trials Steering Committee. They develop and review key messages and documents for both patients and healthcare professionals, including for the Precision-Panc website and other informational material.
There is little doubt that as we understand more about the diversity of disease, we will need to respond to it. Whilst we continue to define similarities and differences between and within cancer types to develop novel therapeutic strategies, platforms such as Precision-Panc, informed by patient input, enabled by precision oncology laboratories and delivered through healthcare systems, will become increasingly important. Such platforms provide practical and efficient ways forward for therapeutic development through aligning discovery, preclinical and clinical development to define lines of sight for novel therapeutic strategies.
Professor Andrew V. Biankin
University of Glasgow (Wolfson Wohl Cancer Research Centre)
+44 (0)141 330 7621